THROMBOSIS RESEARCH 30; 27-45, 1983 0049-3848/83/070027-19$03.00/O Printed in the USA. Copyright (c) 1983 Pergamon Press Ltd. All rights reserved.
ACTIVATOR AND THROMBOPLASTIN SHEEP ALVEOLAR MACROPHAGES
Safia Wasi, Clement E. Burrowes, John 8. Hay and Henry Z. Movat Division of Experimental Rrthology, Department of Pathology, University of Toronto, Medical Sciences Building, Toronto, Ontario, Canada MSS lA8 (Received
17.7.1982; in revised form 5.12.1982. Accepted by Editor R. Reuter)
Alveolar lavage cells from normal sheep were found to be composed of When the cells were cultured, fibrinolytic over 95% macrophages. and thromboplastin-like activities could be detected within 2-4 hours of incubation. As the number of cultured cells was increased the two activities in the conditioned medium increased proportionately. The cells were separated into two distinct subpopulations by means of a sedimentation velocity cell fractionation technique. The macrophage subpopulations were examined for differences in size, morphology, esterase staining and ability to release plasminogen activator and proThese activities were confined to the coagulant activity respectively. The fibrinolytic activity was shown to be large cell subpopulation. plasminogen-dependent and could be inhibited by DFP. On the basis of this the fibrinolytic activity has been designated as plasminogen activator. The procoagulant activity was shown to be thromboplastin in nature because it was Factor VII dependent, inactivated by phospholipase C and not inhibited by DFP. The procoagulant activity has been designated as The two activities could be distinguished on macrophage thromboplastin. the basis of DFP inhibition. INTRODUCTION
Recent studies have indicated that there are substantial differences in macrophage function amongst different species (l-3) and even amongst macrophages from different sites within the same animal (4,s). Moreover, it appears that the macrophages from the same site am heterogeneous in terms of size distribution, enzyme contents and macrophage functions (6-8). Since alveolar macrophages play a major role in the control of infectious processes and in the response to environmental agents, it is important to understand the characteristics of these cells and to determine whether alveolar macrophages are also heterogeneous.
Key words: fibrinolysis.
macrophages, plasminogen activator,
FIBRINOLYSIS, CLOTTING, MACROPHAGES
Our present understanding of macrophage function stems primarily from the studies which employ peritoneal exudate, splenic macrophages and peripheral blood monocytes. Less information is available concerning the defensive and/or immunologic functions of the alveolar macrophages. Alveolar macrophages constitute the largest cell population in lung lavage fluid. They exhibit the geneml characteristics of macrophages (9). They are known to possess receptors for C3b and the Fc portion of the IgG-molecule. They exhibit bactericidal and phagocytic activities. They show appropriate responses to opsoniring antibodies
and to macrophage inhibition factors. They release chemotactic factors which In addition, a distinctive feature of alveolar macroattract polymorphonuclear leukocytes. phages is their ability to adapt to environmental changes, which is reflected by high bio-
synthetic capacity of their endoplasmic reticulum (9). For example, studies on macrophages obtained from cigarette smokers have shown development of arylhydrocarbon hydrolase activity in these cells (9). The spectrum of secretory products of alveolar
macrophages is similar to those of macro-
phages from other sources.
there are no available
tig the presence or release of plasminogen actimtor macrophages in any species.
reports demonstmfrom alveolar
In the skin fibrin deposition has been studied extensively in experimentally induced inflammatory lesions (10, 11) and in allergic contact dermatitis in man (12, 13). Fibrinolysis is important in maintaining homeostasis, since pretreatment of experimental animals with fibrinolytic agents or anticoagulants can, at least partially, prevent fibrin deposition and the formation of inflammatory lesions and indumtion (14-18), although the exact mechanisms of activation of coagulation and of fibrinolysis are not clear. It is well known that in a number of pulmonary inflammatory lesions in man there is both deposition of fibrin and fibrinolysis taking place. Here again the underlying mechanisms are not well understood and hence studies concerned with both thromboplastin and plasminogen activator
In recent years, considemble
macrophages are justified.
on the production
of plasminogen activator and procoagulant activity by a number of inflammatory cells. The procoagulant activity increased greatly when these cells were stimulated in vitro by a variety of stimuli, such as C3b, immune complexes, lectins, ionophores and endotoxin (19-24) and after in vivo exposure of animals to endotoxin (25,26). Similarly plasminogen activator releasencreased considerably when inflammatory cells are stimulated in vitro by agents such as phorbol myristate acetate, asbestos fibers or concanavalin A and in vivo by inflammatory agents such as thioglycollate, endotoxin or mineml oil (27). Furthermore, recently we have shown that inflammatory cells traversing a Freund’s adjuvant induced granuloma can be further stimulated in vivo, to produce higher than basal levels of plasminogen activator, by injecting antigens such as purified protein derivative, directly into the lesion (28). In monocytes and macrophages the relationship between the states of activation of these cells and the capacity to synthesize procoagulant and plasminogen activator activities is not well established. In the studies reported here lavage cells from normal sheep lung were characterized in terms of their morphology, size, estemse staining and capacity to release fibrinolytic and procoagulant activities on fibrin plates in vitro. Evidence is presented which indicates that the plasminogen activator and procxnt activity are rnacrophage-derived mediators.
FTBRINOLYSIS, CLOTTING, MACROPHAGES
MATERIALS Endotoxin Detroit,
055: B5 Westphal preparation)
METHODS was obtained from Difco Laboratories,
Randomly bred sheep (30-40 kilograms) were sacrificed with an intravenous overdose of With the aid of a tracheal tube, the lungs were washed in situ with sodium pentobarbital. heparinized saline, followed by heparin-free saline and finally resuspended in E-s basal medium, without serum. The yield was regularly 2-3 X 108 cells and usually more than 95% were mononuclear phagocytes as documented by non-specific estetuse staining. Differential counting of Leishman stained preparations confirmed that the morphology of more than 95% of the cells was that of mononuclear phagocytes (Figs. 1 and 2). Cell counts were performed with a model B Coulter Counter (Hialeah, FLA) using a 100 micron aperture tube. A standardized window setting for large and small cells was used. Cytocentrifuge smears were stained for non-specific estemse activity according to the method described by Yam et al. (29). Viability of the cells was determined by trypan blue exclusion and over 90%=
The T251-fibrin plates were prepared as described by Wasi et al. (28). The purified human fibrinogen (95% clottable)was mdiolabelled at O°C by the method of MacFarlane (30). The mdiolabelled fibrinogen was diluted with cold human fibrinogen in phosphate buffered saline (PBS) to give 100 pg protein/l00 pl, containing 100,000 to 120,000 cpm. Multiwell tinbro tissue culture plates (35 mm) were coated with 100 pl of this T251-fibrinogen. Prior to use the fibrinogen was converted to fibrin by incubating each well for one hour with 5% fetal bovine serum (FBS) in Eagle’s medium at 37OC in an atmosphere of 5% Cq. Following this procedure, each well was washed three times with PBS and used within 24 hours. Each well contained 80,000 to 100,000 cpm which could be solubilized by For monitoring the lasminogen activator and procoagulant activity, the 0.25% trypsin. varying numbers of washed cells (10 !?3 -108) in 1 ml of Eagle’s basal medium, without serum, were cultured directly in an atmosphere of 5% [email protected]
coated plates under identical
plates for various lengths of time at 3PC on T25I-fibrin In some experiments the washed cells were cultured on unconditions.
In the plasminogen activator
assays the reaction was initiated
by adding 50 t+~ of two
times affinity purified human plasminogen (28) and the plates were incubated at 37’C in an atmosphere of 5% CQ. The fibrinolytic activity was quantitated by withdmwing 25-50 pl aliquots from each well at appropriate times. Soluble 1251-fibrin degradation products were estimated by mdioactivit measurements in an Intertechnique gamma spectrometer (CG30). The amount of TKi_f’b I rin solubilized is expressed as the percentage of the total mdioactivity that can be released by an excess of trypsin. The conditioned medium was centrifuged at 630 g for 20 minutes at 4°C and the supernatant used for further experimentation. Assays for fibrinolytic activity in the conditioned medium were carried out on the 125T-fibrin plates in a final volume of 1 ml of Medium 199 At the end of without serum. The plates were incubated for 15 to 18 hours at 37°C. the test incubation, the radioactivity solubilized in 0.1 ml of the clear digest fluids was counted in a gamma spectrometer. All assays were performed in duplicate with appropriate controls for medium and other reagents. Concentrated conditioned medium from the lung lavage cells was treated with diisopropylfluorophosphate (DFP, 5 X 1O-2M f inal concentration) for 4 hours at 4°C. DFP-
FIBRINOLYSJS, CLOTTING, MACROPHAGES
FIG. Smear of alveolar
lavage cells stained with Leishman’s stain.
of ceils have the morphological
treated and untreated controls were dialyzed exhaustively fibrinolytic and procoagulant activities as described.
The majority X 860.
against PBS and assayed for
Kinetics of release of plasminogen activator by lung lavage ceils were studied by setting up multiple cultures of 1 X 106 cells per ml on ‘25I-fibrin plates as described before. Cell-free supernatants were collected at 2 hour intervals up to 24 hours and assayed for plasminogen activator activity as described. Ml2
were prepared from human serum according
the method of Pitlick and Nemerson (31). Thromboplastin activity was measured by the two-stage assay of Emerson (32). All reagents were pre-warmed and the assay performed at 37°C. In the first stage of the assay a mixture of 100 PI cell suspension or lysate and 100 pl &ICI2 eluate (Factor VII-X concentrate) was incubated for 5 minutes. Then 50 pl of 25 mM CaCl2 was added and incubated for 1 minute. The first stage was stopped by adding 190 CJ of the reaction mixture to 400 Al sodium citmte (15 mM). In the second stage 100 PI of the stage 1 reaction mixture was ad&d to 100 PI citmted pooled sheep plasma and after incubation for 15 seconds, 100 pl of 0.1% lecithin (Centrolex P, Central Soya Chemurgy Division, Port Wayne, IN) was added. Clotting was initiated by the oddition of 100 PI of 25 mM 6Cl2. Clot formation was measured from this point. All assays were performed in triplicate. For comparison dilution of
FIBRINOLYSTS, CLOTTING, MACROPHAGES
The pellet was processed as Section of pellet of alveolar lawge cells. described in MATERIALS AND METHODS and 1 ~MI thick sections were stained with Azure-Eosin. Most of the cells exhibit the nuclear and cytoplasmic
standard prepamtion of commercial were assayed identically.
Cells were fmctionated by unit gmvity sedimentation as described by Miller and Philips (33). Cell suspensions (40 ml) containing 2 X 103 cells were applied on a linear gmdient of l-2% BSA in Hank’s balanced salt solution &IRS) produced from 600 ml of Cells were allowed to settle 1% bovine serum albumin (BSA) and 600 ml of 2% SA. Twenty-three for 4 hours at 4°C and the chamber emptied at a rate of 40 ml/hour. Cells (1 X 106) from each fraction were culfractions of 50 ml each were collected. tured on l251-fibrin plates for 8 hours in the presence and absence of 50 t+ plasminogen. The conditioned media from plasminogen-free cultures were assayed for thromboplastin activity. The plasminogen-dependent fibrinolytic activity was determined as before (28). Snail aliquots were withdmwn at various times and counted for 1251-fibrinopeptides. The effect of the conditioned medium on various deficient plasmas was studied by a onestage clotting assay as follows: 100 pl each of various deficient human plasmas and conditioned medium were incubated for 2 minutes at 37°C and 100 pl of 0.1% lecithin (Centro-
FIBRINOLYSIS, CLOTTING, MACROPHAGES
Fibrinolysis by alveolar lavage cells. Washed cells (1 X 106) were cultured wells in 1 ml Eagle’s basal medium, without serum, in on l251-fibrin-coated Fibrinolysis was estimated by withthe presence of 50 pg of plasminogen. drawing aliquots of medium at various times and counting for mdioactivity in solubilized is expressed as a gamma spectrometer. The amount of l25l-fibrin the percentage of the total radioactivity
that can be released by an excess of
Normal human lex P) was added. Clotting was initiated with 100 pl of 25 mM CaCl2. plasma or buffer (100 mM Tris-HCI, 100 mM NaCI, pH 7.4) were substituted for the deficient plasmas in control assays. The effect of phospholipase C on the procoagulant activity of the conditioned medium Various concentrations of enzyme was studied as described by Gtness et al. (34, 35). (Phospholipase C, B cereus, specificaity 161 units/mg protein, CalbiochemBehring Corp., La Jolla, CAlwere incubated with the conditioned medium for 5 minutes, followed by the clotting assay. The CaCl2 concentmtion was 30 mM.
1 x 106 lung lavage cells
i?i 40i! f 30.t=: .I>
A.__,eA-Al 40 50
A I 60
of lung lavage cells: effect of lasminogen concentmtion. wells in 1 Washed cells (1 X [email protected]
)were cultured on l&-fibrin-coated ml of Eagle’s basal medium, without serum, in the presence of various The reaction was allowed to proceed concentmtions of plasminogen. for 4 hours and aliquots were analyzed for radioactivity us described. The amount of l25I-fibrin solubilired is expressed as the percentage of the total radioactivity
that can be released by an excess of trypsin.
To further chanrcterize the nature of procoagulant activity in the conditioned medium, the latter was subjected to differential centrifugation. Cells (3 X 107 in 3 ml Eagle’s basal medium) were incubated at 37oC for 4 hours with and without endotoxin (30 MB). After incubation the conditioned media were centrifuged at 630 g for 20 minutes at 4°C.
FTBRTNOLYSTS, CLOTTING, MACROPHAGES
g s : S .-iz l-
$ 0 0
/ Number of Cells Added
Increasing numbers of cells (l$-lO*)were incubated on ‘251-fibrin plates for 4 hours and conditioned media were assayed for plasminogen activator and thromboplastin activities as described in MATERIALS AND METHODS. Controls consisted of incubated medium without cells, which gave a clotting time of 142 seconds and less than 5% of the total radioactivity that can be
released b an excess of trypsin; @ represents plasminogen activator vity of 10 ? cells without plasminogen.
The pellets were set aside and the supernatants were centrifuged at 100,000 g for 60 minutes at 4’C in a Beckman Model L5-65 centrifuge, using a SW 50. 1 rotor. The 630 ond the 100,000 coagulant activity
g pellets and their respective supernatants were assayed for proby the two-stage clotting assay.
Smears were prepared from each fraction and stained with Leishman’s stain and for nonspecific esterase. Cell pellets were fixed by suspension in universal fixative (glutamldehyde and formaldehyde) embedded in hydroxyethyl methacrylate and 1 tm sections were cut on a JR4 Sorvall ultmmicrotone. The sections were stained with Azure-Eosin. RESULTS brison
of sheer, alveolar
Based upon the analysis of the lavage fluid
FIBRINOLYSTS, CLOTTING, MACROPHAGRS
Kinetics of release of plasminogen activator by alveolar macrophages. Multiple cultures were established at a concentration of 1 X 106 cells/ ml at 37°C in 5% CO2 in Eagle’s basal medium without serum in the presence of 50-w of plasminogen. The cell-free supernatants were collected at two-hour intervals upto hours and assayed for plasminogen activator activity as described in MATERIALS AND METHODS. The amount of 125I-fibrin of the total radioactivity
solubilized is expressed as the percentage that can be released by an excess of trypsin.
from five sheep there was an average of 95% macrophages, 4% lymphocytes and 1% polymorphonuclear leukocytes and platelets were not evident in the smears or sections (Figs. 1 and 2). This comparison is quite similar to values reported for human lung lavage (9). The fibrinolytic activity of washed lavage Fibrinolysis by cultured lung lavage cells. cells was measured after culturina on l251-fibrin elates. The fibrinolysis Increased progressively as a function of time (Fig. 3). All of the activity de&ted from the lung lavage macrophages was pIasminagen-dependent, showing that it was due to the production of a plasminogen activator. Fibrinolysis from 106 cells per ml was also studied as
FIBRTNOLYSIS, CLOTTING, MACROPHAGES
5 x 2.5
0$ 40 - p 2.0 i C
.z .1. 30z$
0 1.5.c5 u
a T . 0 Fraction Number
Unit gravity sedimentation profile of 2 X 108 sheep alveolar lavage cells. ual of cells (1 X 106)f rom each fraction were cultured for 8 hours on i*“B,Ifibrin-coated wells in 1 ml of Eagle’s basal medium without serum; in the presence and absence of plasminogen. The two-stage clotting assays were performed on conditioned media from plasminogen-free cultures and the results expressed as the reciprocal of the clotting time in seconds X 100. Plasminogen-dependent fibrinoIytlc activity was estimated as described in MATERIALS AND METHODS and the results are expressed as the percentage of the total radioactivity that can be The cell fractions were also stained for nonreleased by an excess of trypsin. specific estemse activity. Results are plotted as percent estemse positive cells in each fraction. numbers
FIBRTNOLYSIS, CLOTTING, MACROPRAGES
a function of plasminogen concentration. As the concentmtion of plasminogen was increased, higher levels of activity were detected from the same number of macrophages (Fig. 4). In the absence of plasminogen even the highest number of cells (107/ml) gave only background activity (Fig. 5). The procoagulant activity of unfructiomted lavage cells. When lo7 cells were cultured on fibrin plates for various lengths of time and the conditioned media were assayed, for thromboplastin activity by Nemerson’s two-stage assay, the cells shortened the clotting time to 61.87 23.51 (SEM) seconds compared to o 164.60 ,+5.82(SEM) seconds in the controls, which consisted of Eagle’s basal medium incubated on fibrin plates without cells. in This increase in activity persisted up to8 hours. The presence of endotoxin (10 &ml> the cell cultures slightly enhanced the thromboplostin-like activity (clotting time of seconds compared to 60 22.01 SEM seconds without endotoxin). 36.7521.05 SEM The conditioned medium from macrophages adherent to plastic plates (non-fibrin-coated) gave identical results for both fibrinolytic and procoagulant activities. Effect of cell numbers on plasminogen activator and thromboplastin release. To study the release of plasminogen activator and thrombpolastin as a function of cell concentration an increasing number of cells (ll.%108 per well per ml) were cultured on l25I-fibrin plates for 4-5 hours and the conditioned media were assayed for the two activities as described. The results obtained are shown in Figure 5. As the number of cells per well was increased the plasminogen activator levels increased proportionately. In the absence of plasminogen even the highest concentmtion of cells was &void of the activity. The clotting times also shortened progressively as a function of cell concentration, indicating a direct relationship between the number of cells and the amount of thromboplastin activity released.
. Kinetics of release of plasminogen activator by alveolar macrophages. When these cells were cultured for various lengths of time and the conditioned media assayed for plasminogen activator activity, the release of plasminogen activator was essentially completed by 8 hours after the cultures were established (Fig. 6). The diminished activity after 10 hours was a consistent finding and may have been due to a negative feedback inhibition of release of activity or perhaps an increased degradation of the plasminogen activator by the proteolytic enzymes secreted by these cells. Similar results were obtained when rabbit peritoneal exudate cells conditioned media were assayed for phospholipase A2 and hyperemia-inducing activities under identical culture condition (36). Unit gruvity sedimentation analysis. by means of unit gravity sedimentation, lavage cells were fractionated with respect to their size. Total cells and the number of large cells were determined in each frnction by means of a Coulter Counter. Cells (1 X lC+)from each fraction were cultured on 1251,fibrin plates for 8 hours and the conditioned media were assayed for plasminogen activator and thromboplastin activity. Sedimentation and activity profiles are shown in Figure 7. Utilizing this technique the lavage cells were separated into two major components, large cells and small cells. Shortening of the clotting time and plasminogen activator activity coincided with the large cell sedimentation profile. Active fractions also showed positive staining for non-specific esterase activity. Leishman’s staining of the cytocentrifuge smears revealed that more than 98% of these large cells had morphological characteristics of macrophages. These macrophages were considered to be activated since they produced plasminogen activator.
FTBRINOLYSTS, CLOTTING, MACROPHAGES
TABLE I Effect of Thromboplastin
on Normal Deficient
Human Plasma and Various Factor Plasmas
TP TP TP
Factor VII deficient Factor VIII deficient Factor Xi deficient
46.3 20.62 209.3 i2.09 45.6 2 0.47 49.020.70
20.62 ? 0.68
Factor VII deficient Factor VII deficient
_+ 1.54 59.08
Factor VIII Factor VIII
101.0_+0.47 >15 min.
Buffer AMCM Buffer AMCM Buffer AMCM
Factor IX deficient Factor IX deficient
Rabbit brain tissue thromboplastin conditioned
100.3 20.85 >15 min.
(40 pg in 100 PI) or alveolar
medium (100 pl from 2 X 107/ml
samples with the above substrates in the one-stage in MATERIALS AND METHODS.
cells) were used as
TP = Rabbit brain tissue thromboplastin; AMCM I Alveolar macmphage conditioned medium; S.E.M. = Standard ermr of mean; NHP = Normal human plasma.
Nature of the procoagulant activity. The procoagulant activity of macmphages described in these experiments was considered to be tissue thromboplastin since it did not cause shortening of the clotting time of VII-deficient plasma. However, the clotting time of normal and of factors VIII and IX-deficient plasmas was shortened (Table I). The presence of thrombin was ruled out by assaying macmphage-conditioned medium with fibrinogen. When 0.1 ml of the conditioned medium was incubated with 0.2 ml of fibrinogen (1 mg/ml), no fibrin clot was formed. This ruled out the presence of any thmmbin in the conditioned medium. The presence of thmmbin, as well as of factors Xa and Va (in the conditioned medium) were ruled out, since DFP-treated conditioned medium had no effect on the clotting capacity of conditioned medium. This suggests that the procoagulant activity derived from the alveolar macrophages was not due to the presence of a serine estemse. For further evaluation we treated the conditioned medium with various concentmtions of phospholipase C for 5 minutes at 3PC (34,35). This treatment abolished the shortening of the clotting time as compared with non-phospholipase C-
FTBRTNOLYSIS, CLOTTING, MACROPHAGES
of Phosphol @se-C
TP TP TP
20 40 80
AMCM AMCM AMCM
40 80 80
on Thromboplastin Clotting
Time 2S.E.M. Get)
59.321.10 112.0 + 0.40 152.6 _+ 1.00 210.2 22.80 397.3 + 1 .oo 48.2 -+ 1.01 162.3 1.41 327.031.02 436.621.17
Rabbit brain tissue thromboplastin (40 pg in 100 pl) or alveolar macrophage conditioned medium (100 pl from 2 X 107/ml cultured cells) were used as samples and incubated with buffer or various concentrations of phospholipase C, as described in MATERIALS AND METHODS. Sheep plasma served as substrate. TP = Thromboplastin (mbbit bmin); AMCM = Alveolar macrophage conditioned medium; S.E.M. = Standard error of mean.
treated material (Table II). Omission of phospholipids from the assay system gave clotting time close to the buffer control values. DFP treatment of conditioned medium had no effect on the shortening of the clotting time in the two-stage assay system. This rules out the possibility that enzymes, like plasminogen activator (371, which are released due to adherence on fibrin, are involved in the mechanism of generation of procoagulant activity. To partially determine the physico-chemical nature of the procoagulant activity, conditioned media were subjected to differential centrifugation, since it has been proposed that tissue thromboplastin is cell-membrane-derived. The results are shown in Table 111. These data show that most of the procoagulant activity was present in the 630 g supernatant and could be sedimented at 100,000 g. There was some activity in the 630 g pellet, but none in the 100,000 g supernatant. In general, there was more activity if the cells had been grown in the presence of endotoxin (10 &ml). When the material containing the disrupted ceils was assayed, no shortening of the clotting time was observed, presumably due to the presence of an inhibitor in the cytosol. DISCUSSION In these studies we have demonstrated that the amounts of plasminogen activator and procoagulant activities released from a certain alveolar cell population correlate well
FIBRINOLYSTS, CLOTTING, MACROPBAGES
1 X 107cells/ml Sample
Endotoxin With Clotting Time -+ S.E.M.
AMCM 630 g supernatant
_+ 0.40 _+ 0.23
20.1 _+ 0.31 22.621.02
630 g pellet
121 .o io.40
100,000 g supernatant 100,000
31 .O + 0.48
The two-stage clotting assay was performed as described in MATERIALS AND METHODS. in the controls Eagle’s basal medium was incubated without cells, with or without endotoxin and the clotting time was 118.0
AMCM P Alveolar error of mean.
with the macrophage content of that population. Unit gravity sedimentation analysis provided a well characterized and enriched population of large sheep alveolar macrophages. The experimental data suggest that macrophages, essentially devoid of other cell types, were able to release plasminogen activator and a procoagulant activity with the characteristics of thromboplastin. These cells were in an activated state in vivo, presumably due to the continual inhalation of environmental irritants or antigens and their subsequent processing by these cells, thereby eliminating the need for further in vitro stimulation. The plasminogen activator activity from lung lavage cells was released quite soon after establishing the cultures, i.e. 2 hours, whereas the same number of cells in afferent lymph plasma required 15 hours in culture to release equivalent levels of activity (28). This reflects the differences in the macrophage content of these cell populations - i.e. 98% in alveolar lavage fluid and 515% in afferent lymph plasma (28, 38). The five-&y old glycogen induced rabbit peritoneal exudate consists of an avemge of 60% macrophages (36). Comparable levels of plasminogen activator activity is released fram these cells after about 8 hours in culture (wasi - unpublished observation). Hence it appears that the cell populations of various anatomical origins release plasminogen activator activity proportional to the percent of their macrophage contents. From the experiments described here we are unable to determine whether the higher levels of plasminogen activator activity from the lavage cells represent a higher activation state, in addition to a larger proportion of macrophage content
in this cell population.
FIBRINOLYSIS, CLOTTING, MACROPRAGES
The criteria utilized for the characterization of the macrophage procoagulant activity strongly suggest its thromboplastin-like nature. It was assayed by the two-stage assay of Nemerson (32). It could only be detected in the presence of Factor VII. The procoagulant activity was inactivated by phosphol ipase C and was unaffected by MP. It is not known whether these cells had or had not been “stimulated” by lymphocytes The unfmctionated lavage cells contained about 95% in situ, prior to the lawge. macrophages and about 4% lymphocytes, by morphologic criteria. Recently, Levy and Edginton (39) have presented evidence for collabomtion between murine splenic macrophages and lymphocytes for the amplification of the macrophage procoagulant activity. No increase in activity could be recovered from separated populations of lymphocytes Unstimulated lymphocytes were also or macrophages when stimulated by endotoxin. unable to trigger the increase in the macrophage proccugulant activity. Only lymphocytes which were pre-exposed to endotoxin were able to amplify macrophage procoaguUsing human peripheral blood cells, Edwards et al. (40) demonstrated lant activity. that adherent monocytes produced tissue factor-like activityFn stimulation with endotoxin and that non-adherent lymphocytes enhanced this activity. These findings may not necessarily be appl icable to alveolar macrophages. Exposure of the lung lavage macrophage cultures to endotoxin suppressed all soluble plasminogen activator activity. Chapman et al. (41) have shown that hioglycollate-stimulated mouse macrophages in medium containing endotoxin synthesize an inhibitor of plasminogen activator activity. We have also detected a plasminogen activator inhibitor activity in the conditioned medium from the PPD (purified protein derivative of tuberculin) stimulated afferent lymph macrophages. Partially purified inhibitor blocked the urokimse mediated fibrinolysis as well as kallikrein action on benzyl arginine ethyl ester Wasi, unpublished observations). The cellular effect of endotoxin in an inflammatory environment appears to be more in favour of fibrin deposition. On sedimentation velocity sepamtion of alveolar lawge cells the plasminogen activator profile was almost identical with the thromboplastin profile (Fig. 7). In other words the same population of large cells seems to be responsible for both the procoagulant and the fibrinolytic activity. This would suggest microlocal macrophage heterogeneity, des(8), in the same cell population. There could be seveml explanacribed by Suga et. tions for this microlocal heterogeneity. (a) The production of different secretory products of macrophages in response to the same stimulus may occur at different rates and at different times. (b) Adaptive mediator production may take place in response to unevenly distributed stimulating component(s) existing in microlocal foci within an organ, a region or an inflammatory lesion. (c) Adaptive mediator production may also tuke place in response to other cell types (gmnulocytes, lymphocytes, plasma cells, endothelial cells and fibroblasts) present in the vicinity of an activated macrophage. The numbers and the variety of these cells may vary microlocally. The products released by these cells may govern the rates and times of production of mediators. (d> An activated macrophage producing thromboplastin may have emigmted from elsewhere to the alveolar space and lodged itself adjacent to a residont plasminogen activator producing macrophoge or vice versa .
At the moment we are unable to detect whether the same cell is producing the wo activities or the two adjacent macrophages (activated simultaneously) are producing the two opposing activities. single cell assays should be devised to help resolve this question. However, from the data on the kinetics of release of the h~o activities it appears that the
FIBRINOLYSIS, CLOTTING, MACROPBAGES
first of the above I isted explanations may account for this microlocal heterogeneity. The isolated lavage macrophages do not start releasing plasminogen activator instantly. The enzyme release begins after a few hours (present study) to 24 hours (28, 37,42) in culture. However, the procoagulant activity is demonstrable by the sheep alveolar lavage cells (present study), as well as sheep afferent lymph cells (Hay, unpublished observations) prior to culturing. There appears to be precedence for this sequence of events. The genemtion of thromboplastin will initiate the extrinsic coagulation pathway, giving rise to fibrin formation on the cell surface. This event would immobilize the activated macrophages at the inflammatory site until they start secreting plasminogen activator and presumably inhibitors of blood coagulation. This will lead to the solubilization of the fibrin network around the cell and migmtion of the macrophages from the inflammatory site. This sequence of events has an in vivo counterpart. In lobar pneumonia fibrin deposition an early phenomenon, whereas resolution - i.e. fibrinolysis is a late occurrence.
It has been established in a number of mediator systems that there is first, synthesis, secretion or release of mediators followed by the appearance of restricting factors. Pathological processes develop when there is either excess production or insufficient restriction. ACKNOWLEDGEMENTS The studies reported were supported by the Medical Research Council of Canada (Grant MT-N% to J. 8. H.); the Ontario Thomcic Society to (H.Z.M.) and the Ontario Heart Foundation
is a Career Investigator
The authors wish to thank Mr. Warren Chin for Medical Research Council of Canada. his assistance with the unit gravity sedimentation and Ms. Marina Michael for expert secretarial
BOETCHER, D.A. Description
and MELTZER, M.S.
DAUBER, J.H. macrophages.
M ouse mononuclear cell chemotaxis: Inst. 54,795-799, 1975.
and DANIELE, R.P. Secretion of chemotaxin by guinea pig lung I. The spectrum of inflammatory cell responses. Exptl. Lung Res.
Chemotaxis of mononuclear
TERRITO, M.C. 187-192, 1977.
TERRITO, M.C. and GOLDE, D. W. The function J.Reticuloendothelial Sot. 25, 11 l-120, 1979.
of human alveolar
WEINBERG, D.S., FISHMAN, M. and VEIT, B.C. peritoneal macrophages. I. Effector cell activity geneic and xenogeneic tumor cells. CeII.Immunol.
Functional heterogeneity among of macrophages against syn38,94-104, 1978.
FISHMAN, M. and WEINBERG, D.S. Functioml heterogeneity among peritoneal macrophages. II. Enzyme content of macrophage subpopulations. 611 .ImmunoI. 45_ 437445, 1979.
FIBRINOLYSIS, CLOTTING, MACROPHAGES
SUGA, M., DANNENBERG, A.M. and HIGUCHI, S. Macrophage functional Identiheterogeneity in vivo: Macrolocal and microlocal macrophage activation. fication by double-staining tissue sections of BCG granulomas for pairs of enzymes. Amer. J. Pathol. 99,305-324, 1980. BERNARD,
GEE, L. and FICK JR., R. B.
R. B. and DVORAK,
Role of the clotting
Kinetics of fibrinogen/fibrin
system in cell-mediated
and cutaneous basophil hypersensitivity
R. B. and DVORAK, H.F. Fibrinogen-fibrin on the surface of macrophages: detection, distribution, binding requirements and possible role in macroJ.Ex&&d. i42,1377-1390,1975. phage adherence phenomem. DVORAK,
R. B., JOHNSON,
hypersensitivity. permeability changes in tuberculin
system in cell-mediated hypersensitivity. I. l&686-698, skin reactions in man. J.Exp.Med.
Fibrin deposition 1973.
The effect of heparin on the ocular tuberculin and BICK, M.W. Arch. Ophtha lmol . 61, 709-711, 1959.
Effects of anticoagulants and other drugs on cellular and cutaNELSON, D.S. neous reactions to antigen in guinea pigs with delayed-type hypersensitivity. immunology,~ 219-234, 1%5.
S. B., BENACERRAF,
R.T. and OVARY,
H. J. and LESKOWITZ,
The effect of carrageemn E,
FEINMAN, L., COHEN, S. and BECKER, E. L. The effect of fumaropimaric acid on delayed hypersensitivity and cutaneous Forssman reactions in the J.1 mmunol. 104,1401-1405,197O. guinea pig. PRYDZ,
complement 1977. 20.
and SCHORLEMMER, H.-U. Further link between Nature, (Land. 1270, 173-l 74,
and blood coagulation.
J. H. Leukocyte procoagulant activity. by igE and antigen-antibody complexes.
SPIEGELBERG, H.L. and VAUGHAN, Enhancement of production in vitro J.Clin.invest.
FJBRTNOLYSTS, CLOTTING, MACROWAGES
of tissue thromboplastin complexes and lectins. 22.
RIVERS, R. P.A.,
Thrombos. Haemostas. 39, HATHAWAY,
duced procoagulant 316,1975. 24.
P. and ALLISON,
of human monocytes.
1978. W. L. The endotoxin-in&it. J. Haematol. 30,311-
Tissue factor generation by human mononuclear cells: dissociation of tissue factor generation from mitogenic Med. 89,792-883, 1977. 25.
In vitro stimulation
(factor TIT) activity in human monocytes by immune Thrombos.Res. 3465-474, 1979.
PRYDZ, H. and ALLISON, monocytes.
effects of endotoxin and response. J.lab.Clin.
Tissue factor activity.
J. and FANI,
REICH, E. Activation of plasminogen: a general mechanism for producing localized extracellular proteolysis. In: Molecular Basis of Biological Degmdative Processes. R.D. Berlin, H. Herman, I.H. Lepow and J.M. Tanzer (Eds.) New York-San Francisco-London: Academic Press, 1978, pp. 155-169.
production of plasminogen activator by afferent but not efferent lymph ccl Is 128; emigrating from chronic granulomatous lesions in sheep. J.Immunol. 1076-1082, 1982. 29.
L. T., LI, C. H. and CROSBY,
monocytes and gra nul ocytes. 30.
MacFARLANE, l82,53, 1958.
factor apoprotein. In: San Fmncisco-London:
45, L. Lomnd (Ed. > New YorkMethods in Enzymology Academic Press, 1976, ‘pp. 37-48.
NEMERSON, Y. The phospholipid requirement tion. J.Clin.Tnvest. 47, 72-80, 1%8.
and PHILIPS, R.A.
of proteins with
Amer. J. Cl in. Pathol . 55,283-290,
Se paration 1%9.
of tissue factor
in blood coagula-
of cells by velocity
FTBRTNOLYSTS, CLOTTTNG, MACROPHAGES
E. and BERRE, A.
from Bclcillus cereus and its use in studies of tissue thromboplastin. Biochem. 27,238-243,1972. 35.
OTNESS, A-B., LITTLE, C., SLETTEN, K., WALLIN, R., JOHNSON, S., FLENGSNUD, R. and PRYDZ, H. Some characteristics of phospholipase C from Bacillus
Eur. J. Biochem. 79,459~468,
lipase A2 mediates
amplification 1244, 40.
H. Z. and HAY, J. B. Extmcellular hypememia. Nature, =,583-585,
J.C. and COHN, Z.A. Induction of macrophage by endotoxin stimulation and phagocytosis: Evidence for J.Exp.Med. 140, 995-1010, 1974.
J. B., PEDERSEN,
in inflammatory 39.
plasminogen activator a two-stage process. 38.
and MORRIS, B. The role of lymphatic Ser. Haemat. Vol. III, 2, 17-61, 1970.
T.S. Lymphocyte cooperation is required for of macrophages procoagulant activity. J.Exp.Med. 151, 1232-
EDWARDS, tissue factor:
cell of origin and requirements
Blood, 54, 359-
370, 1979. 41.
CHAPMAN, H.A., VAVRIN, Z. and HIBBS, J. B. Macrophage fibrinolytic activity: identification of two pathways of plasmin formation by intact cells and of a plasminogen activator inhibitor. gl,28,653-662, 1982. KLIMETZEK, gen activator
V. and SORG,
C. Lymphokine-induced secretion of plasminoby murine macrophages. Eur. J.Tmmunol. 3 185-187, 1977.